PC Experiencing Random Shutdowns (Possibly Overheating)

The4rchitect

Honorable
Jan 22, 2014
43
0
10,540
Hey guys,

My PC will shut off within five minutes from boot randomly, no "Windows has encountered an error" message or anything. Just turns right off. When I try to turn it back on, I'll get a beep, the fans will start up, and after about five seconds it will turn off again, and I'll have to give it an hour before trying again (which will just give me five minutes).

In the limited time I have to look around before it shuts down, I've noted HWMonitor has my CPU temp climbing steadily up to 89*C (yes, Celsius) and then it always shuts down right as it is about to hit 90*C.

I've had the PC for about a 16 months, custom built. I did a check of all my fans to make sure they're running, cleaned out the dust, and removed and re-applied thermal paste (correctly) to the CPU. I believe it's an overheating issue and not a power supply or motherboard issue as can sometimes be the case, but I don't know for sure and would like some help/advice on what to do next. Thank you for your help, specs below!

AMD FX-8350 4.0GHz 8-Core Processor
Corsair H55 57.0 CFM Liquid CPU Cooler
Gigabyte GA-990FXA-UD3 ATX AM3+ Motherboard
Crucial Ballistix Sport 16GB (2 x 8GB) DDR3-1600 Memory
Crucial M500 240GB 2.5" Solid State Drive
Seagate Barracuda 1TB 3.5" 7200RPM Internal Hard Drive
XFX Radeon R9 290 4GB Double Dissipation Video Card
NZXT Phantom 410 (Gunmetal/Black) ATX Mid Tower Case
XFX 650W 80+ Bronze Certified ATX Power Supply
Samsung SH-224DB/BEBE DVD/CD Writer
Microsoft Windows 8.1 (64-bit)
 

Hex-

Commendable
Feb 18, 2016
4
0
1,510
I also have an r9 290 and an 8350. I get those random shutdowns during gaming and bluescreens with the errors CLOCK_WATCHDOG_TIMEOUT and WHEA_UNCORRECTABLE_ERROR.

Run Memtest and OCCT (do OCCT first, run memtest over night). OCCT should help you determine if it's thermal or voltage related.
 

Golfis

Honorable
Aug 31, 2013
75
0
10,660
Well since you have already found out the source of your issue, i'm not sure we can do anything about it. Make sure that the cpu cooler is properly mounted, and whether or not the pump is actually working. While starting up, should make a slight noise while the coolant starts to flow through the tubes.
 

The4rchitect

Honorable
Jan 22, 2014
43
0
10,540
Hey guys, sorry for the delay in response, wanted to try a few things before posting.

I was going to borrow a friend's PC and swap out parts to try and locate the issue but ended up not.

I took a second look at my CPU cooler, the Corsair H55, I wasn't hearing the coolant flow as Golfis suggested so I thought I'd located the problem.

I moved one of the 3 pin connectors from its spot at "CPU_FAN" to another empty spot at "SYS_FAN" and, surprise, the coolant started to flow when I powered up.

Now, I have no idea why this would happen, as somewhere along the way it stopped working, but I DID NOT move any cables, yet when I moved the cable this time it starts to work again. Any explanation for this? It would make sense if I had moved the connector before, and then moved it back, but this is the first time I've changed connections so that doesn't make sense.

Regardless, I thought this would fix my temps, but CPU temps still start climbing rapidly towards 90*C. Before, it would ALWAYS shut down as it was about to go to 90*C, but this time the temp kept climbing, stabilizing slightly at around 100*C for about 20 minutes of uptime. Some light touching of the tubing holding the coolant had it feel warm to the touch, but not even close to how I would imagine it would feel were boiling water flowing through them.

At this point I'm sure I need a BIOS update, so I check my updates to find I'm running a very old version (F2) when Gigabyte is on F12. I've never updated the BIOS and thought it happened automatically.

So, I've downloaded the new BIOS update but haven't installed, as I've heard if I lose power in the middle of flashing my BIOS I'll brick my PC. I'm glad I waited, because about 30 minutes into my session my PC shut off again, when the CPU temp ran about 115*C.

Restarting it saw the same issue, turns on for a second then shuts right off. SOMETHING had to happen for me to start reaching longer uptimes, but I haven't fixed my issue yet. I'm hesitant to update the BIOS in case it shuts off mid-update.

What do I do now? Thanks for the help and I'll respond quickly this time!
 

The4rchitect

Honorable
Jan 22, 2014
43
0
10,540
Any help? Would like to avoid taking it to a computer shop if possible, if I do and they update the BIOS and my PC ends up getting bricked would I be covered for that? I'm out of all my warranties.

Is there a safe way to update my BIOS to prevent the possibility of bricking?
 

Golfis

Honorable
Aug 31, 2013
75
0
10,660
Unfortunately i do not think a bios update will solve any of your problems. Bios updates are meant to improve compatibility, not solving hardware issues. Is your cpu, cooler and motherboard under warranty? if yes then try to send them for a repair, if not, then try to find another heatsink (air cooled), the cheapest one you can get, mount it to your motherboard and check the temperatures. As for the cables, something might be sorting out and ends up causing these issues. Also the water pump needs 12 V to operate normally, if you carefully read the manual, it will probably state that you are not meant to operate the pump with anything lower than 12V as you might end up damaging it.
 

The4rchitect

Honorable
Jan 22, 2014
43
0
10,540
Ok, so I found my stock cooler that came with the FX-8350 finally and installed it, I'm currently 20 minutes into watching my temps at idle and they float between 30*C and 50*C, with the stock cooler running at 3000-4600 RPM (the stock cooler sounds pretty loud, but that's not a problem I can replace it once I know the cooler is the issue).

I guess I should expect the temps to be slightly lower with a better cooler, should it be getting up to 50*C at idle? I'll make another post later when I try out a game or something, don't want to push it right now.

I'll probably let it sit for a while and go through my BIOS updates, I'm certain I'm getting accurate readings as HWMonitor and AMD OverDrive are both giving me the same numbers. That should be accurate then, right?

 

Golfis

Honorable
Aug 31, 2013
75
0
10,660
Whoahh, that is quite a big gap in temperatures... Do you mean that simultaneously , some cores operate at 30 *C , and some others at 50*C ? What is the room temperature? Also take a look at the task manager to see if something is using the processing power of the cpu in the background, thus keeping it active
 

The4rchitect

Honorable
Jan 22, 2014
43
0
10,540
These are CPU temps, it does look like a lot of stuff is running, normally helper programs for Adobe Photoshop, AutoCAD, etc. I disabled all the startup stuff but I always have a bunch of minor processes running at idle